Theories of early cooperation in human society often draw from a small sample of ethnographic studies of surviving populations of hunter–gatherers, most of which are now sedentary. Borneo hunter–gatherers (Punan, Penan) have seldom figured in comparative research because of a decades-old controversy about whether they are the descendants of farmers who adopted a hunting and gathering way of life. In 2018 we began an ethnographic study of a group of still-nomadic hunter–gatherers who call themselves Punan Batu (Cave Punan). Our genetic analysis clearly indicates that they are very unlikely to be the descendants of neighbouring agriculturalists. They also preserve a song language that is unrelated to other languages of Borneo. Dispersed travelling groups of Punan Batu with fluid membership use message sticks to stay in contact, co-operate and share resources as they journey between rock shelters and forest camps. Message sticks were once widespread among nomadic Punan in Borneo, but have largely disappeared in sedentary Punan villages. Thus the small community of Punan Batu offers a rare glimpse of a hunting and gathering way of life that was once widespread in the forests of Borneo, where prosocial behaviour extended beyond the face-to-face community, facilitating successful collective adaptation to the diverse resources of Borneo's forests.

Susan Keates and Yaroslav Kuzmin have contributed valuable comments on our assessment of the chronology and technological characteristics of Shuidonggou localities 1 and 2. These comments have demonstrated some discordance in our publications that should be corrected here. Their rationale for abandoning the conclusions altogether is, however, weak at best. Most importantly, there is no reason to return to the chronology for the Initial Upper Palaeolithic (hereafter termed IUP) assemblages at the Shuidonggou site complex (SDG) proposed by Madsen et al. (2001). Keates and Kuzmin's critique focuses on two separate issues: the first is our characterisation of the assemblages from the lower part of SDG 2, and the second is our handling of the dates. We will address these in order.

This paper examines choices of earth-working tools made by Neolithic Chinese populations. In the Hemudu Culture (7000–5000 B.P.), bone (scapula) digging tools were used from the earliest times, whereas peoples in surrounding areas used stone spades. A range of experiments on manufacturing costs, durability, and use efficiency under realistic conditions show that bone and stone spades are functionally equivalent when soils are soft, but that stone implements provide significant and easily perceived advantages when working harder soils. The persistence of scapular spades in the Hemudu Culture would have constrained decisions about undertaking large construction projects under normal soil conditions. Our results show that, in addition to generalized labor for construction, labor demands for producing earth-working implements for large-scale prehistoric earthworks could have also been substantial. These findings not only help explain the processes of intensifying rice-agriculture and sedentary settlements in the Lower Yangzi Basin, but also create a solid foundation for further investigation of how the recruitment of both generalized and specialized laborers, the organization of craft production, and the relevant logistics for large-scale earthworks may have paralleled concentrations of political power in prehistory.

The Shuidonggou site cluster in northern China contains 12 different early prehistoric sequences with great potential to cast light on the transition to Upper Palaeolithic behaviour in East Asia. Here researchers present the latest results from Locality 2, reporting seven occupation levels with hearths, animal bone and diverse industries. Although previously compared with European Upper Palaeolithic sequences, the new work proposes a different trajectory of development. Distinctive macroblade technology arrived in the area, possibly from Mongolia or Siberia, about 41000–34000 years ago. This industry subsequently disappeared, to be replaced by flake technologies.

The earliest Upper Palaeolithic industries of the Levant, which figure prominently in discussions of the spread of anatomically modern humans and the origins of the Upper Palaeolithic, are known from a small number of localities. Two sites in the Hatay region of Turkey have yielded initial Upper Palaeolithic assemblages similar to those found in the Levant. One of the sites, Üçaǧizlı’ cave, has also provided radiometric dates and faunal remains, both relatively rare for sites of this period.

Toby Morrow argues that I (Kuhn 1994) made an error in calculating the utility/mass ratio of retouched tools, and he proposes an “alternative” formula. Unfortunately, Morrow confounds utility per se with ratio of utility to weight, and, as a result, his alternative approach is erroneous. This response to Morrow examines the central issue of how artifact utility is assessed. The main message of my original paper is reaffirmed. Utility/mass ratios eventually begin to decline as artifacts get large, and optimal size ranges can be estimated based on a few simplifying assumptions.

It is widely believed that the design of transported artifacts and toolkits employed by mobile populations is influenced by two main factors, portability and potential utility. Choices about the kinds of artifacts to carry around can be modeled as an optimization problem, in which it is beneficial to obtain the greatest potential utility for the minimum total weight. This study approaches the problem analytically, making a few simple assumptions about artifact geometry and the relations between utility and artifact size. If artifact utility is calculated as a function of potential for renewal, transported toolkits should consist entirely of relatively small finished tools. Moreover, most gains in durability or multifunctionality that require increases in overall size are outweighed by increased transport cost. Inconsistencies between these expectations and ethnographic and archaeological observations point to situations in which artifact functionality is more closely constrained by overall size or mass.

Variation in the forms of Mousterian tools has been attributed to a number of factors, including style, function, and more recently, differential reduction. This study compares the influences of two factors, reduction and the shapes of tool blanks, on the forms of scrapers from a Mousterian site in Italy. In the subject assemblages, the shapes of the tool blanks had a much stronger influence on scraper forms than had reduction. To the extent that blank form affects typology in other cases, explanations of contrasts among Mousterian assemblages would need to account for the use of alternative techniques of flake production. Since typological variation is multicausal, it would also appear more profitable to focus directly on variables pertinent to current research issues, rather than on reinterpreting the typology.